Personal Style Learning in Sumi-e Stroke-based Rendering by Inverse Reinforcement Learning

スポンサーリンク

概要

• 論文の詳細を見る

• We consider the problem of automatically generating sumi-e style drawings using machine learning techniques. In our previous work, we regarded a brush as a computer agent and trained the brush to generate smooth strokes to fill given boundaries under a pre-designed cost function. In this paper, we extend this approach and propose to also learn the cost function from a user's real brush stroke data by inverse reinforcement learning. This extension allows the brush agent to imitate the personal drawing style of a user. The effectiveness of our method is demonstrated through experiments.

• 一般社団法人情報処理学会の論文
• 2013-11-21

著者

• Ning Xie

  Department Of Computer Science Tokyo Institute Of Technology

• Masashi Sugiyama

  Tokyo Institute of Technology

• Ning Xie

  Tokyo Institute of Technology

• Tingting Zhao

  Tokyo Institute of Technology

関連論文

• Artist Agent A2: Stroke Painterly Rendering Based on Reinforcement Learning
• Personal Style Learning in Sumi-e Stroke-based Rendering by Inverse Reinforcement Learning
• Personal Style Learning in Sumi-e Stroke-based Rendering by Inverse Reinforcement Learning

スポンサーリンク